Review of Abstract: "Improving the representation of major Indian crops in the Community Land Model version 5.0 (CLM5) using site-scale crop data"

Title: Improving the representation of major Indian crops in the Community Land Model version 5.0 (CLM5) using site-scale crop data. This is a fitting title that points directly to the nature of the work done in the study.

Introduction and Purpose:

The abstract establishes the significance of accurate cropland representation in terrestrial simulations in CLM, effectively focusing on spring wheat and rice, which dominate agricultural land use in India. The introduction is clear and compelling, providing a strong rationale for the study. The impetus for the study is critical and timely as accurate representation of crop functional types in non-temperate regions of the world is an essential and current research concern in many Land Surface, Earth System, and Dynamic Global Vegetation Models (LSMs, ESMs & DGVMs) - largely due to data paucity issues. Improving the accuracy of the representation of cropland in a reputable DGVM like CLM will therefore contribute to the field of cropland and plant functional type representation in DGVMs overall. As a side note, it may be useful to mention a key "error" with the previous state of crop modeling in CLM that the study now addresses.

Methodology:

The methodology is compelling, as it outlines the creation and utilization of a novel, comprehensive spring wheat and rice database to improve the parameterization of crop phenology, growing season, and simulated yield. The use of eight sites encompassing 20 growing seasons for each crop hints at the robustness of the study. The abstract elucidates how this data was used to calibrate the relevant CLM5 crop functional types, situating the improvements achieved and mentioned in the results, in the appropriate methodological context.

Results:

Results outline specific enhancements to CLM5's performance metrics for simulated crop functional types, with comparisons to alternative datasets (MODIS). Significant improvements in Pearson’s r values for various simulated crop features like LAI, GPP, and corresponding energy fluxes provide good evidence that the study objectives were effectively achieved, demonstrating the improved accuracy of the model.

Conclusion:

The impact of the study is clear, accentuating the need for region-specific crop functional type parameterization in global LSMs, ESMs, and DGVMs. Broader implications for modeling land-atmosphere interactions across various climate scenarios add value to the research.

Overall Assessment:

This is a comprehensive abstract that effectively outlines the purpose, methodology, results, and conclusion of the study. It maintains an appropriate balance between brevity and detail, making the abstract both informative and accessible.

Suggestions for Improvement:

The abstract could however further clarify the novelty of the dataset that was used. Additional detail on aspects of the data that make it unique or unprecedented would be valuable, in the context of the relevance of the data to the study. Secondly, the abstract could benefit from an additional sentence (or two) that emphasizes the broader implications of the study, addressing for example, how the improvements made can be CLM use in practical contexts like climate impact modeling on agricultural land. This will add to the utilitarian relevance of the study. Thirdly, a brief mention of any challenges or limitations (e.g., calibration process or data digitization issues) of the improved model would provide a more rounded perspective of the outcomes of the study. 

Summary:

This abstract effectively communicates the significance, methods, and results of the study, making a compelling case for the improved CLM5 model with the work that was done. It could be even more impactful with the minor adaptations mentioned above if the authors deem the recommendations highlighted useful. The resulting paper would be one to look forward to.

General Comments:

Marked improvements for spring wheat and rice in India using the CLM5 land model, largely achieved through the calibration of key crop growth and planting parameters. The growing seasons now better align with observations, addressing the previous errors in the crop calendar. Useful and important work. The introduction of latitudinal variation in base temperature is a valuable and important addition.

Specific comments

• L124: Would adding a daylength control on planting date / crop emergence help here ?
• Why did the 0.4 grain fill threshold for rice perform poorly, while a 0.65 threshold showed improved results? By making this change, you are effectively decreasing yields and growth. How do you justify that a 0.4 threshold worked well at other sites (original value), but a 0.65 threshold yields better results at the studied sites? This seems to connect with the paragraph on Line 439; it would be interesting to expand on this further.
• Figure 3, simulated LAI during early season growth is generally much higher than observed, why is this ? Also, are there different Spring Wheat cultivars between sites, this could influence results, would be interesting to include this if relevant.
• 1.1.2 Yield – is it possible to separate grain yield from biomass growth, this would be an interesting distinction to make here for the site validation.  
• By drawing on the mean yield (t/ha) across sites, simulated vs. observed, you might be masking the model's strengths and weaknesses. A more transparent way of illustrating this performance metric would be beneficial. For example, the way you have illustrated LAI validation clearly shows that some sites are better simulated than others, which is normal and important to note.
• In figure 6 another neat way of illustrating this would be the spatial differences between obs and sim yields. – likewise for figure 8 and GPP. Visually this could aid the interpretation of results. 
• Line 434 this is a very good point, in further work it could be interesting to see whether it is possible to include the option of multiple rice harvests in one year (where agriculturally feasible) in CLM ?
• With monthly time series of sensible and latent heat fluxes you are essentially capturing how well the model captures seasonality, to dig deeper into how well these fluxes are simulated it would be beneficial to uncover weekly if not daily fluxes.

Technical corrections:

• Change “Site data used in validation” to “Site data used for validation” (in the supplement).
• Omit “the” in Line 37.
• In Line 125, “planting temperature” is repeated twice; delete one instance.
• I would personally omit the code illustration in Lines 180 to 188 and keep the code in the supplementary material; a description in the main body suffices.
• In Line 496, when you mention that they are off by at least three months, it would be useful to specify whether the peak LAI by CLM5_Def is early or late.

Review of “Improving the representation of major Indian crops in the Community Land Model version 5.0 (CLM5) using site-scale crop data, K. Narender Reddy, Somnath Baidya Roy, Sam S. Rabin, Danica L. Lombardozzi, Gudimetla Venkateswara Varma, Ruchira Biswas, and Devavat Chiru Naik”

Summary

The article focuses on improving the representation of two major Indian crops, spring wheat and rice, in the Community Land Model version 5.0 (CLM5) to enhance its accuracy in simulating crop phenology, yield, and associated land-atmosphere interactions. Using a newly created, comprehensive site-scale crop data set from India, the study calibrated and adjusted key parameters in the CLM5 crop module, such as sowing dates, growth parameters, and base temperature. The modified model versions (CLM5_Mod1 and CLM5_Mod2) demonstrated significant improvements in simulating crop growth, water and carbon fluxes, and irrigation patterns compared to the default CLM5 version. These modifications underscore the importance of region-specific parameters for global land models and provide a basis for better understanding land surface processes and their role in climate scenarios. The study's findings have implications for regional agricultural management and policy, as well as for enhancing climate modeling accuracy.

Title

The title generally works well with the content of the manuscript but mentioning the specific crops worked on in the title would provide readers with some clarity on what to expect.

Abstract

The abstract provides a concise summary of the study's goals, methodology, and findings, stating that the modified CLM5 performs better in simulating crop phenology, yield, and fluxes. For instance, it mentions that the Pearson’s r for monthly LAI improved from 0.35 to 0.92 and monthly GPP from -0.46 to 0.79 compared to MODIS data.

While the abstract states that it aims to improve the representation of Indian crops, it could be more specific earlier on by naming the two crops (spring wheat and rice). This would immediately inform the reader about the study's focus. Consider revising the sentence: "Our study aimed to improve the representation of these crops in CLM5" to "Our study aimed to improve the representation of spring wheat and rice in CLM5."

The abstract could briefly mention the broader implications of these improvements. For example, it could say, "These improvements can enhance the accuracy of land-atmosphere interaction studies and inform regional agricultural management and policy."

Introduction

The introduction effectively outlines the importance of accurately simulating cropland processes in Land Surface Models (LSMs), which impact energy, water, and carbon fluxes. It provides sufficient background on the Community Land Model (CLM) and its development up to version 5.0.

While the introduction cites several relevant studies (e.g., Elliott et al., 2015; Lombardozzi et al., 2020), it could benefit from a few more recent references to highlight the current state of crop modeling. For example, "Recent studies provide valuable insights for enhancing the accuracy of simulating biogeophysical and biogeochemical processes..." could include more studies published after 2020 to strengthen this point.

The introduction mentions that "CLM5 simulations of rice and wheat over the Indian subcontinent show large biases," but it could be clearer about what specific biases (e.g., underestimation of yield, incorrect phenology timing) the current study addresses. Adding a sentence such as, "Specifically, the model has been shown to inaccurately simulate the timing of planting and harvesting for spring wheat and rice, leading to incorrect estimates of carbon fluxes and water use," would provide a clearer problem statement.

Materials and Methods

The description of the CLM5 model and its modifications (CLM5_Mod1 and CLM5_Mod2) is comprehensive, outlining the data sources, simulation setups, and parameter modifications. The distinction between site-scale and regional-scale simulations is also clearly made.

The detailed methodological description would benefit from a flowchart or diagram summarizing the process, from data collection to model calibration and evaluation. For example, Figure 1 in the document effectively shows the sites used for calibration, but a flowchart could visually represent the steps outlined in Sections 2.1 to 2.3.1.

The manuscript describes various parameter changes (e.g., base temperature, planting dates), but it could provide more justification for selecting these specific parameters for sensitivity analysis. For example, the section "Improvements in CLM5" states, "The base temperature and maximum GDD control the longevity of each phase in crop growth," but it does not explain why these were chosen over other potential parameters. A brief explanation could be added, such as "These parameters were chosen based on their significant influence on phenological development stages in crops, as indicated by previous studies (cite studies)."

Results

The results are presented with clear visualizations, such as the Taylor diagrams and time-series plots, that compare model versions against observational data.

While the Taylor diagrams (e.g., Figure 4) effectively show improvements in the model's performance, they could benefit from a brief explanation of how to interpret them. For example, "Higher correlation, lower RMS error, and smaller standard deviation characterize the most accurate CLM5 configuration, as seen in the closer proximity of CLM5_Mod2 markers to the observational reference point."

The results section presents the remaining biases in yield and growing season length (e.g., "The growing season length simulated by CLM5_Def is very low with a mean growing season of just 69 days, compared to 129 days in observations"), but it could discuss potential reasons for these biases in more detail. For example, it could mention model assumptions, data limitations, or unaccounted-for environmental factors that could contribute to these discrepancies.

Discussion

The discussion appropriately links the results to the study objectives, emphasizing the importance of region-specific parameters in LSMs for improving crop simulation accuracy.

The manuscript mentions, "Such improved land models will be a great asset in investigating global and regional-scale land-atmosphere interactions and developing future climate scenarios," but it could expand on specific applications. For instance, how could this model be used to inform irrigation management practices or forecast agricultural productivity under different climate scenarios?

The discussion would benefit from a dedicated section on limitations and future directions. For example, the text could state, "While the modified models showed significant improvements, there are still biases that could be addressed by incorporating more diverse site data or accounting for multi-cropping practices in the model," and suggest specific future research that could address these limitations.

Conclusion

The conclusion reiterates the key findings and reinforces the need for region-specific model calibration.

The conclusion could be strengthened by including a call for further studies or actions, such as "Future work could focus on extending this modeling approach to other major crops in India or integrating socio-economic factors to better inform policy-making."

Figures and Tables

Figures and tables are generally well-presented and labeled, effectively supporting the text.

Consider adding more comparative visuals that summarize the improvements across different metrics and model versions. For instance, a bar chart comparing MAB, RMSE, and Pearson’s r values for CLM5_Def, CLM5_Mod1, and CLM5_Mod2 could provide a quick visual reference for readers.

References

The references are relevant and extensive, covering a range of studies on LSMs and crop modeling.

Include more recent references (post-2020) to ensure the manuscript reflects the latest advancements in the field. For example, search for recent studies on crop modeling in LSMs that may have incorporated new methodologies or datasets.

General Comments

The manuscript is generally well-written with a clear technical style suitable for the target audience. However, some sections could benefit from simplified language to increase accessibility for readers from diverse scientific backgrounds.

The manuscript follows a logical flow, but the Methods and Discussion sections could be further refined for clarity and depth.

Final Recommendations

Provide more justification for the choice of specific parameters in the sensitivity analysis and model modifications.

Include a discussion on the potential policy and practical implications of the findings and suggest specific areas for future research.

Consider adding more comparative figures and diagrams to succinctly showcase the differences between model versions and their performance improvements.

By addressing these critiques, the manuscript can be strengthened to ensure a robust and impactful contribution to the field. If you need further assistance with more detailed critiques on specific sections or figures, let me know!